Fault Identification of New Energy Based on Online Monitoring

Online monitoring and fault diagnosis technology can reduce maintenance costs, for improve the operational reliability have great value. With the rapid development of new energy technologies, this paper takes the wind turbine pitch system as the research object. Used the supervisory control and data acquisition(SCADA) system as experimental data, which stored in the database. The SCADA system dates presence of large number of similar samples, So in this paper eliminate the redundant information data based on the similarity function. Then using the method of nonlinear state assessment techniques established the pitch system NSET health model use the normal operation dates. when the device in normal operation, the residuals between the actual values and the predict of the model within the threshold range. when the system in abnormal operation, the residuals between the actual value and the predicted value of the model exceeds the threshold range. The actual verification shows that the NSET model can accurately identification the fault, thereby improved the new energy of safety and economy.